.

Background

Surface chlorophyll has been in decline with increasing warming scenarios, and several - controversial results have also come in this field
Recurring blooms in the North Eastern Arabian Sea and associated with subsurface - chlorophyll maxima(SCM)
Stratifications in NEAS are strengthening in NEAS as well as globally
Remote Sensing data underestimates column primary production

Objectives of the study

Critical Questions to address

What are the biophysical interactions of phytoplankton and fluid dynamics in SCM layers ?
How to monitor the extent of distribution of Chl-a with changes in potential density on synoptic levels
How to describe bloom dynamics in physical terms?
How to track the changes in frequency domains of Chl-a in response to climate change
How to assess the forecasting effects and challenges in predicting

Chl-a dynamics

SCM

Typical chlorophyll profiles

Why SCM is Important ?

It is a prominent feature in tropical seas.
Have significant contribution in PP
But mostly it is underestimated by currently using PP models.
Exhibit strong seasonality
A Better indicator of climate change

What are the variations and its patterns?

Strong inverse correlation with Mixed Layer Depth(MLD)
Atmospheric forcing and stability of water column are important factors which drives its occurrence.
Phytoplankton species composition is a major factor on determining the peak magnitudes.
Species composition is positively correlated with PAR intensity

Timeline of major studies in Marine chlorophyll dynamics

Study location, Algorithms and sampling

Observation & Methods

Study location, Algorithms and sampling

Data Sources

Bio Argo profiles[2013-2018]
In situ CTD profiles[2013-2018]
Wind data (ASCAT)
Aqua MODIS PAR data.
Aqua MODIS Fluorescence
Tropoflux SWR
Published datasets

CTD spec & Probes

SBE (Model 11 PLUS, Sea-Bird Inc.) in 1 m bin intervals.
The CTD system is equipped with auxiliary sensors of dissolved oxygen, fluorescence and photosynthetic active radiation (PAR)

Study Location

The study region North Eastern Arabian Sea is monitored using both CTD and BioArgo
Season- quiescent and warm month of Spring InterMonsoon
Preferred sampling from open ocean to find the innate upper ocean dynamics due to stratification

Algorithms & programing

All the Algorithms, Analytics, Automations , Empirical modeling , spatial maps, simulations and documentations are done using R programming Language

Clustering and sorting SCMs according to seasons

Fitting with Non-Linear Gaussian curves

An inverted gaussian can easily represent the SCM

Rasterizing the fit values

Surface-subsurface chlorophyll distribution and comparison

Data strategy for temporal analysis

Understanding Process and Interactions

Phytoplankton and Fluid dynamics

Understanding Process and Interactions

Seasonality of PAR on SCM

Seasonality of PAR affects surface chlorophyll
But SCM didn’t have much effects on PAR
The PAR have a stable exponential decay at subsurface

Seasonality of PAR on SCM

Seasonality of PAR affects surface chlorophyll
But SCM didn’t have much effects on PAR
The PAR have a stable exponential decay at subsurface

How Fluid density drives the chlorophyll?

Fluid Simulations on 10 most abundant species

Four species were more focused due to its abundance in this location concurrent to BioArgo

Noctiluca scintillans
Guinardia striata
Thalassiothrix longissima
Nitzschia longissima

Stable stratified conditions favours settling of phytoplankton

In stable sea conditions (March/spring) particle sinking or phytoplankton settlement can be well explained by Stokes Theorem
Phytoplankton maintains the buoyancy by its innate mechanisms of gas vacuoles and lipids and often are heavier than seawater
The density of phytoplankton depends on the volume rather than mass which depends on respiration and photosynthesis

- This disproportion in mass and volume makes the phytoplankton (especially diatoms) positively buoyant after a critical size range (Gross and Zeuthen, 1948)

Density-settling velocity relationships very strong in March

A simple experiment

This experiment is analogous to the sinking phytoplankton groups
The two distinct density layers created by oil and water
The sinking of mustard seeds are very related to the sinking of green Noctiluca
The same mechanism can be expected in the NEAS

Physical processes associated with Green Noctiluca Blooms

Surface-Subsurface oscillations of Chlorophyll-a

Physical processes associated with Green Noctiluca Blooms

Changes in Bloom dynamics

Blooms have positive relations with hot weather
The current warming scenario in NEAS is not different , and shows increasing yearly.
The NEAS system shows a significant community shift (Diatom to Dinoflagellate) from the year 2000
The addition of mixotrophic dinoflagellate to system enables early blooms and enhance the productivity in the location

How blooms changes in NEAS

When it blooms

Influence of extreme events

But it is entirely absent in early years

The role of surface currents

The observations found that, surface currents have influence in lifting phytoplankton groups to surface waters. This is well evident from Empirical Orthogonal Teleconnections.
The lifing is more sensitive when startified waters with shallow MLDs are prominent

Spatial signatures of Chl-a

Wind

{.r-stretch}

Blooms and Surface-Subsurface interactions

Trophic dynamics of SubSurface layers

The LVM models suggest that the Trophic dynamics are efficient in energy transfer
The nutrient dynamics on the subsurface are stable
The temperature and light factors are optimum
The subsurface blooms can be expected more frequent in future warming scenarios

Redefining Sevrdrup Model

\[ H_c \approx h_l \frac{\mu_0}{m} \]

Here \(H_c\) is critical depth, \(\mu_0\) is the phytoplankton population growth rate at the surface in the absence of any biomass loss, \(h_l\) is the light extinction coefficient and \(m\) is the biomass loss due to respiration, zooplankton grazing, viral lysis, and mortality which is assumed to be in constant depth.

Considering density (Sigma-t) rather than depth(critical depth)

According to the current study, any layer of strong density gradient in the upper layer is favorrable for phytoplankton \(d\sigma t_{max}\) , as it results in low settling velocity, which is expressed as follows

\[ d\sigma t_{max} \approx [P_z.T_z] \times \frac{Chl_z}{Vz} \]

\[ Chl_z = \frac{d\sigma_{max}}{P_z.T_z} \times V_z \]

Global density changes, settling velocities and time shifts in blooms

Phytoplankton response to climate change

Global density changes, settling velocities and time shifts in blooms

SST variations in decadel scale over four weeks of March

Times series analysis shows an increasing trend throughout the years
The Anomalies are occurred due to extreme events such as ENSO
It is found that the first week of March end with the temperature more than the occurred in the last weeks in its intial phase

SCM strengthen and warming seasons on various years

Units

Temp.surf - \(\circ\) C

Sigma.SCM - \(kg/{m}^3\)

Warming effects on Noctiluca distributions

Warming -Stratification links to Noctiluca are well reflected in spatial distributions
The Northern Hemisphere seems to be subjected to more than south
The Land locked areas are more prone to the occurrence of Noctiluca blooms due to nutrient availability and efficient stratification

Wavelet analysis of blooms in NEAS at WM-SIM transition

Early startification leads to early blooms

The chlorophyll values were high(early onset blooms) only during March in the earlier days
However, the onset of blooming is observed to be early in the recent years
The Month of February and March is very critical for blooms, due to the withdrawal of convective mixing and initiation of stratification during the period
As reflected in the changes in density, it is concluded that the early blooming is due to the early stratification

Surface chlorophyll relation with pycnocline upper layers

wavelet decomposs of Pycnocline depth and chlorophyll(add color)

The Wavelet decompositions suggest that pycnocline depth and surface chlorophyll have interactions
The low frequency decompositions of pycnocline are well aligned with surface chlorophyll

Wavelets are strongly aligned in D1 & D2

The wavelet decomposition in low frequencies (D7 and D8 waves) are strongly concurrent between the surface chl-a and pycnocline depth
The periods seems to be frequent within a range of 128 & 256 days
The changes in pycnocline are significant in NEAS because the nutrient -depleted surface waters enrich by pycnocline due to wind mixing

Wavelet Coherence: pycnocline depths vs surface chlorophyll

Wavelet Coherence: pycnocline vs SCM

Potential density anomalies over the decades(change color)

potential density vs extreme events

Potential density Anomaly- \(kg/m^3\)

How SCM depth is controlled by SST-PDEN ratio

Application of SST-PDEN relations

The SST-PDEN ratio relationship with SCM is utlized to derive Biologically Influenced Stratification Index (BISI)
BISI describes the probable SCM in locations, where the SST is increasing and density decreases

Limitations, possibilities and Applications

Forecasting submerged Chlorophyll with Machine Learning

Limitations, possibilities and Applications

Model simulations & sensitivity

\[ Chl_{max} = \frac{\sigma \times (-1.23)+ par \times (-0.0044) + 30.41 }{{{ {[{\frac{-D-D_{max}}{Chl_{spd}}}]}^2 }}} \]

Positive changes in chlorophyll by physical state of water

Positive changes in density of water strongly associated with productivity in NEAS
This density changes are mostly considered as a cue of nutrient rich water
But the advances in density can also considered as a buoyancy support for phytoplankton at surface

Changes on physical state is visible throughout the year

Neural Nets for deep learning and prediction

Efficiency of Neural Nets

Chlorophyll dynamics is Stochastic or Chaotic ?

Chaos & Chl-a

In a chaotic system an initial condition always leads to the same final state, following a fixed rule, while in a stochastic system, an initial condition leads to a variety of possible final states

Nature of chaotic dynamics

it must be sensitive to initial conditions,
it must be topologically transitive,
it must have dense periodic orbits.

Chaos Test

In the regular case (periodic or quasiperiodic dynamics) the trajectories for the system are typically bounded
In the chaotic case the trajectories are typically behave approximately like a two-dimensional Brownian motion with zero drift and hence evolve diffusively

The Biological variables seems chaotic

The biological variables such as chlorophyll and oxygen are observed to be chaotic
Chaotic variables are unpredictable with regular statistical approaches
But can be traceable with adequate long range datasets

The SST in SIM have slipping dynamics from regular to chaos
The SST in WM is on the track
The salinity data seems very stable on SIM and WM

The density changes are a function of sst and salinity
The slippery effects on SST also reflected in the density at SIM

Summary and Conclusion

Interactions of phytoplankton with prevailing stratified fluid systems are highly reflected in the vertical distribution of chlorophyll profiles

The insights reveals that the warming have immediate impacts in phytoplankton groups, by increasing its settling velocity resulting in subsurface aggregation

Settling in the subsurface leads to increase the plasticity among phytoplankton which thrive in low light

As global warming is an emerging threat, the findings will direct towards low light abundance as a critical wayout defining the adaptation strategies of the phytoplankton community

Summary and Conclusion

The subsurface is comparatively suitable in the warming scenario, than surface, in terms of stability, nutrient availability and optimum temperature

The warming scenarios have varying influences on local and global scales, and the gradient in surface density is getting decreased over the period due to stratification

Stratification results in strengthening of new productive subsurface, the optimum layer for several phytoplankton species, and maintain as a prominent subsurface chlorophyll maxima (SCM)

Stratification in oceanic systems with proximity to landlocked areas are favourable for mixotrophic Noctiluca (blooming), which have been regular since 2000

Summary and Conclusion

Seasonal stratifications is observed to be early in the recent years, resulting in bloom formation in advance especially in landlocked areas, available with nutrients.

Both fluid and phytoplankton innate mechanisms (motility) associated with complex ecosystem interactions create chaos, and so is difficult to forecast the submerged chlorophyll.

Estimates of primary production is gives lower values without the submerged chlorophylls

Deep learning algorithms such as Artificial Neural Nets can be best utilized in predicting chlorophyll profile